Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for performing a region growing comprising: identifying a first seed voxel; comparing a Hounsfield unit value of voxels adjacent to the first seed voxel with a threshold Hounsfield unit value for air; identifying the adjacent voxels as seed voxels, when each of the adjacent voxels meets or exceeds the threshold Hounsfield unit value for air; determining whether the identified adjacent voxels are a leakage object; segmenting newly identified adjacent voxels, which are determined not to be a leakage object; iterating the identifying, determining, and segmenting steps; determining whether a new adjacent voxel is segmented in a previous iteration; determining whether a leakage object is identified in the previous iteration when no new adjacent voxel is determined to have been segmented in the previous iteration, wherein leakage is determined based on (a) an inversely proportional relationship of a front size of adjacent voxels and an iteration number and (b) a linear relationship between the front size and bifurcations and changes in an airway shape; and reducing the threshold Hounsfield unit value and performing the region growing algorithm only on the identified adjacent voxels, as seed voxels, in the previous iteration when at least one leakage object is determined being identified in the previous iteration, wherein the method is performed by a processor.
A method, performed by a processor, for growing a region in a 3D image to create a virtual model of a branched structure like an airway. The method starts with an initial "seed voxel" and compares the Hounsfield unit value (density) of neighboring voxels to a threshold value representing air. Voxels meeting or exceeding this air threshold are added as new seed voxels. The algorithm checks if these new voxels represent a "leakage object," determined by a combination of the size of the voxel front (inversely proportional to the number of iterations) and airway shape changes (linear relationship between front size and bifurcations). Voxels determined not to be leakage are segmented. This process iterates, and if leakage is detected, the air threshold is lowered, and the region growing continues only from the previously identified adjacent voxels.
2. The method according to claim 1 , further comprising terminating the region growing process when no leakage object is identified in the previous iteration.
The region growing method from the previous description includes a step to terminate the process when no leakage object has been identified in the immediately preceding iteration. This stopping condition prevents over-segmentation and ensures that the region growing accurately represents the branched structure, automatically ending the process when the algorithm is not finding additional relevant voxels and when leaks are no longer being detected.
3. The method of claim 1 , wherein the first seed voxel is a point in a trachea.
The region growing method described previously, used to create a virtual model of a branched structure, starts with the first "seed voxel" being specifically located within the trachea. This provides a consistent starting point for the region growing process, particularly when modeling the human airway, allowing the algorithm to accurately segment the lungs and other connected structures from a defined anatomical location.
4. The method of claim 3 , wherein the first seed voxel in the trachea meets a high threshold value.
The region growing method, starting with a seed voxel in the trachea, as previously described, requires that this initial seed voxel also meets a high Hounsfield unit threshold value. This ensures that the starting point for the segmentation process is robust and minimizes the chance of starting the region growing within a low-density area that is not part of the targeted anatomical structure, contributing to a more accurate model.
5. The method of claim 1 , further comprising counting a number of the segmented voxels in each iteration.
The region growing method described previously, which creates a virtual model, further includes counting the number of segmented voxels during each iteration of the algorithm. This provides quantitative information about the progress of the region growing and can be used to monitor the segmentation process, detect potential issues, and assist in the "leakage object" determination logic by observing the rate of segmentation.
6. The method of claim 5 , wherein the identified adjacent voxels are determined to be a leakage object when an increase in the number of segmented voxels exceeds a predefined rate.
In the region growing method that counts segmented voxels in each iteration, as described previously, a determination that the identified adjacent voxels represent a "leakage object" is made when the increase in the number of segmented voxels between successive iterations exceeds a predefined rate. This rate-based detection prevents the algorithm from erroneously continuing segmentation into areas outside of the anatomical structure being modeled.
7. The method of claim 6 , wherein the segmented voxels are identified and recorded in each iteration.
Within the region growing method that segments voxels and determines "leakage objects" based on a predefined rate, as previously explained, the segmented voxels are identified and recorded in each iteration of the algorithm. This allows for tracking the evolution of the segmented region, providing data for visualization, analysis of segmentation performance, and potential correction of errors in subsequent processing steps.
8. The method of claim 7 , creating a voxel list from the segmented voxels of each iteration.
In the region growing method where segmented voxels are identified and recorded in each iteration, as described previously, a voxel list is created from the segmented voxels accumulated across all iterations. This list represents the final segmented region and can be used for subsequent steps, such as generating a 3D model, measuring volumes, or performing other analytical tasks on the segmented structure.
9. The method of claim 8 , generating a model from the voxel list.
Using the region growing method where a voxel list is created from the segmented voxels of each iteration, as previously described, a 3D model is generated from this compiled voxel list. This enables the visualization and analysis of the segmented anatomical structure. This model can be used for diagnostic purposes, surgical planning, or other applications that require a representation of the segmented region.
10. The method of claim 9 , wherein the model is of an anatomical structure.
The model generated from the voxel list, derived from the region growing algorithm as previously described, specifically represents an anatomical structure. This resulting anatomical model, such as a representation of the lung airways, is used for medical imaging applications, including diagnostics, surgical planning, or research, offering a visual representation of patient-specific anatomy.
Unknown
December 5, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.